Jorge Jovicich wrote:
> ------------------------------------------------------------------------
> Name: MOTION.GIF;1
> MOTION.GIF;1 Type: GIF Image (IMAGE/GIF)
> Encoding: BASE64
>
Dear Jorge,
looking at your realignment estimates, it could be that what the realignment
algorithm
perceives as "motion" is actually of another origin than real head motion.
E.g. changes
in global image intensity (slight variations of the TR?) or more local image
intensities
(due to variations in B0 inhomogeneity / RF instability) that appeared every
second or
third TR.
So look at the voxel intensity changes over time in your images e.g. by
plotting the
global normalization (does this also jitters regularly?). Also check that if
your slices
were acquired in an interleaved way that even and odd slices have +/- the
same
intensities - "cross over" would give a sort of "zebra" appearance on SAG or
COR sections
(if acquisition was TRA).
Can you give us some details about your acquisition (machine, sequence, TR,
interleaved?)
Stefan.
--
Stefan Sunaert, MD
Department of Radiology, KUL MR Research Centre
University Hospital Gasthuisberg, K.U.Leuven
Herestraat 49, B-3000 Leuven (Belgium)
Phone: +32 16 347754, Fax : +32 16 343769
mailto:[log in to unmask]
http://www.kuleuven.ac.be/radiology/Research/People/Stefan/
|